Garage Construction Project

Yard Elevation Survey — DIY Field Procedure

13 min read

Yard Elevation Survey — DIY Field Procedure

What this is

A field-followable procedure for mapping the relative ground elevations of the backyard between the house and garage using tools already on hand — the Bosch GLL50-20 cross-line laser, pine grade stakes, and a 48” aluminum ruler as the measuring rod. The output is a rough contour map (~±½”) that (1) feeds Terry’s drainage plan and (2) — via the invert tie-in in Phase D — answers the utility-trench sewer-slope question.

The method is classic differential leveling: the laser is the “level,” the ruler is the “rod,” and shared “turning-point” stakes carry one datum across each move. This is the detailed how-to for the method summarized in the trench doc.

Day or night? Both — split the work

PhaseWhenWhy
A — Layout + stakesDaytimeDriving stakes, laying out the grid, and writing labels are all far easier in daylight.
B–D — Laser readingsDusk → darkThe GLL50-20 is a cross-line laser with no receiver mode, so outdoors you’re eyeballing a dim red line — it’s only visible once it’s dark enough.

So: lay the whole grid out in the afternoon, then come back at dusk with the laser.

Tools & materials

  • Bosch GLL50-20 cross-line laser + tripod (1/4”-20 mount)
  • 48” aluminum ruler / straightedge (the “story pole” / rod) — run a strip of white tape down one face so the red line reads against the shiny metal
  • ~60 pine grade stakes (17.5”) + mallet
  • Index cards + staple gun (white targets stapled to each stake)
  • Red laser-enhancement glasses (~$10) — biggest single visibility win at night
  • 100’ tape measure (grid layout) + marker (Sharpie) for labels
  • Torpedo level or phone level (optional — to hold the ruler plumb)
  • Clipboard + tally sheet (template below) + phone camera

Phase A — Daytime: lay out the grid & set stakes

Grid precision barely matters — this is an elevation survey, not a boundary survey

The grid is just a pattern of points where you read height. Each elevation (HI − FS) is independent of how straight the rows are, whether spacing is exactly 10’, or which way is “north.” A grid that’s slightly compressed, skewed, or crooked still gives correct elevations — only the plotted map distorts a little, which doesn’t move the high/low spots or the garage-vs-house drop.

  • Forget true north — “rows/columns” are just labels; align them to whatever’s convenient. Your lot probably isn’t square to north anyway, and it doesn’t matter.
  • Use the garage wall as the master straightedge for one axis (you already trust it). A taut mason’s string holds a row straighter than a tape over grass. A neighbor’s fence works as a second straightedge if it’s straight and roughly parallel — but don’t assume it’s the property line or perfectly straight.
  • The one number that must be accurate isn’t the grid: the straight-line run from garage stub to house tie-in (a single careful tape pull), which the sewer-slope calc needs. Grid = contour picture; that tape pull = the run.

Want it square anyway? Check the diagonal

Optional — the heights don’t need it, but it makes a tidy plot and helps if the same layout also guides the trench route or porch/driveway edges.

  • One square corner: the 3-4-5 triangle (mark 3’ one way, 4’ perpendicular, adjust until the hypotenuse = 5’). Use a bigger multiple over a big grid — 30-40-50 beats 3-4-5.
  • Whole rectangle: measure both corner-to-corner diagonals — when they’re equal, it’s square (opposite sides equal). Run the baseline off the garage wall, mark equal widths at both ends, nudge the far corners until the diagonals match.
  • Or compute the target: diagonal = √(L² + W²) — a 60’×40’ grid → √5200 ≈ 72.1'.
  1. Pick the origin and orientation. Choose a corner as your reference (e.g., the garage’s SE corner = column 0, row A). Run the 100’ tape along one edge as the baseline.
  2. Drive the grid on a 10’ spacing. Stakes every 10’ along columns (1, 2, 3…) and rows (A, B, C…). Tighten to 5’ near the back-of-house / porch / driveway edge, where Terry’s grading happens and you want more resolution.
  3. Drive each stake firm. Exposed height does not need to be uniform — you measure to the ground, not the stake top.
  4. Label each stake with its grid ID (A1, B3, C5…) in marker, and staple a white index card near the top as the laser target.
  5. Plan the laser setups. Mark a few setup positions (high spots with good sightlines) and note which stakes will be turning points (TPs) shared between consecutive setups. Aim to grab as many stakes per setup as you can — fewer hops = less accumulated error.
  6. Establish a benchmark (BM). Pick one fixed, re-findable point — the garage slab top is ideal, or drive a dedicated datum stake. Assign it an arbitrary elevation, e.g. 100.00”. Every other point’s elevation will be relative to this.
  7. Sketch the grid on your tally sheet and photograph the layout before dark.

Wind

Stapled index cards stay put far better than taped paper. On a breezy night, that’s the difference between readable targets and chasing flapping paper.

Phase B — Dusk/dark: laser readings (differential leveling)

Flip-calibrate first (2 minutes, do it indoors)

Project the horizontal line on a wall and mark it. Rotate the laser 180° and re-check: the line should land on the same mark (within ~⅛” at that distance). A 5-year-old cross-line laser can drift — confirm it’s still true before you trust a night’s readings.

  1. Set up the laser. Tripod it at the first setup point and let it self-level — keep the tripod roughly level so you stay inside its ±4° self-leveling range (if it blinks/flashes, it’s out of range). Put the red glasses on. The horizontal line is a forward fan, not a 360° plane — rotate the laser body to face each cluster of stakes.
  2. Set the plane height wisely. The line must hit the ruler at every stake in range — so the plane has to sit above the highest ground but no more than 48” above the lowest (or the reading runs off the top of the ruler). On big relief, take a shorter reach / more setups.
  3. Backsight (BS) the benchmark. Hold the ruler plumb with its 0” end on the ground at the BM. Read where the laser line crosses the ruler = BS. Compute the Height of Instrument:

    HI = (benchmark elevation) + BS

  4. Foresight (FS) every stake in range. Ruler plumb, 0” on the ground, read where the line crosses = FS. Then:

    Elevation = HI − FS

    Sign check: a bigger reading means the ground is lower (the level plane sits farther above it).

  5. Read the turning point (FS on the TP stake) and record it before you move — that fixes the TP’s elevation for the next setup.

Phase C — Leapfrog to the next setup

  1. Move and re-level the laser at the next setup position.
  2. Backsight the same turning-point stake you just read: BS₂ on the TP → HI₂ = (TP elevation) + BS₂. Now you’re back on the same datum.
  3. Foresight the new batch of stakes (Elevation = HI₂ − FS). Pick the next TP, read it, and repeat. Keep leapfrogging until every stake is read.
  4. (Accuracy, optional but worth it) Close the loop: at the end, return and re-shoot the original benchmark. The gap between its measured and known (100.00”) elevation is your closure error — small is good; distribute it evenly across the setups if it matters.

Tally sheet template

StationBS (backsight)HIFS (foresight)ElevationNotes
BM (slab)30.0130.0—100.0datum = 100.00
A1—130.042.088.0
C5—130.028.0102.0high spot
D5 (TP)—130.035.095.0turning point → setup 2
D5 (TP)20.0115.0—95.0re-shot from setup 2
E5—115.041.074.0low spot

The math in words

Benchmark = 100.0”. Setup 1 backsight on it = 30” → instrument plane (HI) = 130”. A stake reading 42” is 130 − 42 = 88" (a foot lower than the slab). The turning point D5 reads 35” → 95". Move the laser; from setup 2 the same D5 reads 20” → HI₂ = 95 + 20 = 115". Now a new stake reading 41” is 115 − 41 = 74". Every elevation shares the one datum.

Phase D — Invert tie-in (the sewer-critical step)

The grid maps the dirt surface — but the sewer’s fall is governed by pipe inverts, not the ground. Carry the same datum down to both inverts so the survey answers the real question:

  1. Garage stub invert — from a stake of known elevation near the garage, measure straight down to the 3” PVC slab stub-out invert. (Closes the mech-room sewer invert To-Do.)
  2. House basement stack invert — carry the datum to the house and measure down to the lowest workable tie-in on the stack. A water level (hose level) is simplest here — it carries one datum around the house corner with no line-of-sight needed; use the laser+ruler only if you have a clear shot. (Closes the house sewer-stack tie-in depth To-Do.)
  3. Compare: garage-stub elevation − house-stack elevation = available fall. Required fall = run × ¼”/ft (80’ → 20”; 100’ → 25”). The garage stub must sit at least that much above the house invert.

Surface elevations ≠ inverts

Don’t let a tidy surface contour map talk you out of measuring the stack invert. Even if the two ground surfaces look near-level, the house basement stack sits ~5’ down — that depth is the lever that usually still yields enough fall. The invert numbers decide shared-trench-vs-separate-trench, not the dirt.

Field checklist (quick reference)

  • Day: grid laid out (10’, tighten to 5’ near house), stakes driven + labeled, index cards stapled
  • Day: benchmark chosen + assigned 100.00”; setup positions + turning points marked; layout photographed
  • Dusk: laser flip-calibrated; red glasses on
  • Each setup: tripod roughly level (self-level locked), plane height within 48” reach
  • Backsight the benchmark/turning point → record BS, compute HI
  • Foresight every stake (ruler plumb, 0” on ground) → record FS
  • Read the turning point before each move; re-backsight it after
  • (Optional) close the loop back to the benchmark
  • Phase D: tie in garage stub invert + house stack invert; compare to required fall
  • Photograph each stake (label + elevation) for the docs; transcribe the tally sheet

Common mistakes (and the fix)

  • Ruler leaning → reads long. Hold it flat to the stake, plumb it by eye or with a torpedo level. Always rest the 0” end on soil, not on a grass tuft.
  • Too many hops → error compounds. Set up on high ground, grab lots of stakes per setup, and re-read each turning point twice.
  • Plane too high → reading runs off the 48” ruler. Lower the tripod, or split into more setups.
  • Reading a fat line → take the center of the beam consistently.
  • Trusting surface for the sewer → see the Phase D [!important] callout.

Make it a learning project (doing it with a kid)

This survey is a near-perfect summer science project — it turns abstract math into something you build, measure, and see in the backyard. Here’s how to run it with Conor — an all-A student heading into 6th grade, so this is right in his wheelhouse — and learn together. He doesn’t need trig: a 3-4-5 triangle and some subtraction is plenty, and it’s a fun refresher for the grown-up too.

Jobs he can own

A real survey crew has a “rod person” and a “recorder” — give him real titles and real responsibility:

  • Stake driver / target stapler (daytime) — drives the grid stakes and staples a white index card on each.
  • Square checker — owns the 3-4-5 tape and tells you whether a corner is square (below).
  • Rod person (night) — holds the 48” ruler plumb on each stake and calls out where the red line crosses it. (This is a genuine surveying job — the “rodman.“)
  • Recorder — writes each reading into the tally sheet. Reading numbers off and logging them is half of real fieldwork.
  • Cartographer (after) — colors the finished grid of numbers into a hills-and-valleys map of the yard.

The math, made real (no trig required)

  • Right angles you can hold — the 3-4-5 trick. Measure 3 ft along one string and 4 ft along the other; when the diagonal between the ends is exactly 5 ft, the corner is square. Have Conor check it: 3×3 = 9, 4×4 = 16, 9 + 16 = 25, and 5×5 = 25. He just did the Pythagorean theorem with a tape measure — no formula needed. (Scale it up: 6-8-10, or 30-40-50 for the big grid.)
  • Why the laser line is always flat — gravity. The yard slopes, but the laser (like water in a hose) finds dead level every time. A good “why does that happen?” conversation.
  • Subtraction with a purpose. Every height is just instrument height − the reading. Bigger reading = lower ground. The arithmetic means something because it tells you where the water will run.
  • Turning numbers into a picture. A page of readings becomes a colored map — high spots one color, low spots another. That’s real data visualization, and the payoff is predicting where rain flows (then checking if you were right next storm).
  • “Close enough” is a real idea. Measure one stake twice; the numbers won’t match exactly. A great moment to talk about measurement error and why we say ~±½”.

For the grown-up refresher

The trig under the hood

The 3-4-5 trick is the Pythagorean theorem: a² + b² = c². For a full rectangle, equal diagonals mean square corners, and the diagonal length is √(length² + width²) — e.g., a 60×40 grid → √(3600 + 1600) = √5200 ≈ 72.1'. Dust off the square-root key on your phone and you’re surveying.

Keep it safe (and fun)

Laser eye-safety with a kid

The Bosch GLL50-20 is a Class 2 laser — the natural blink reflex protects against a brief glance, but the rules still matter with a curious 10-year-old: don’t stare into the beam, never look into the laser’s aperture, and don’t aim it at anyone’s face. Set the laser low and pointed at the rod, not at head height where someone could look back into it. Make knowing the rule his job too. See 01 - General Safety Requirements for the project’s eye-protection standard.

A few more for a night in the yard:

  • Headlamps for both of you (hands free for the rod and the clipboard).
  • Watch the footing — the grid stakes are little trip-and-poke hazards in the dark; walk the rows you know.
  • Bug spray, and quit while it’s still fun — better to leave him wanting another night than to grind to the last stake.
  • And of course — the late night is Mom’s call. 😊

Last Updated: June 25, 2026

Graph View

Backlinks